From 4286972a0f8d072328fb6a295964f63e017f76e5 Mon Sep 17 00:00:00 2001
From: Wei Ji <23487320+weiji14@users.noreply.github.com>
Date: Sat, 12 Aug 2023 08:55:38 +1200
Subject: [PATCH] Geopandas pyogrio driver test code

Adding class to load files into a geopandas.GeoDataFrame using pyogrio engine. Supports file formats like geoparquet and flatgeobuf.
---
 h5tests/geopandas_pyogrio_arr_mean.py | 33 +++++++++++++++++++++++++++
 1 file changed, 33 insertions(+)
 create mode 100644 h5tests/geopandas_pyogrio_arr_mean.py

diff --git a/h5tests/geopandas_pyogrio_arr_mean.py b/h5tests/geopandas_pyogrio_arr_mean.py
new file mode 100644
index 0000000..e91dbde
--- /dev/null
+++ b/h5tests/geopandas_pyogrio_arr_mean.py
@@ -0,0 +1,33 @@
+import os
+import subprocess
+
+import h5py
+import numpy as np
+import pandas as pd
+
+from .h5test import H5Test, timer_decorator
+
+try:
+    os.environ["USE_PYGEOS"] = "0"
+    import geopandas as gpd
+    import pyogrio
+except ImportError:
+    completed_process = subprocess.run(
+        ["mamba", "install", "-c", "conda-forge", "pyogrio", "--yes"]
+    )
+    import pyogrio
+
+
+class GeopandasPyogrioArrMean(H5Test):
+    @timer_decorator
+    def run(self):
+        group = "/gt1l/heights"  # not used
+        variable = "h_ph"
+        geodataframes = []
+        for file in self.files:
+            # file = "s3://nasa-cryo-scratch/h5cloud/flatgeobuf/ATL03_20230211164520_08111812_006_01.fgb"
+            # print(f"Loading {file} into geopandas via pyogrio")
+            gdf = gpd.read_file(filename=file, engine="pyogrio")
+            geodataframes.append(gdf[variable])
+        final_geodataframe: gpd.Series = pd.concat(objs=geodataframes, axis="index")
+        return np.mean(final_geodataframe)